Assessment of soil surface roughness statistics for microwave remote sensing applications using a simple photogrammetric acquisition system

In this paper we present a simple and efficient method to measure soil surface roughness in an agricultural environment. With the deployed system one can generate digital surface models (DSMs) with a minimum size of 1 × 2.5 m2 extendable to any desired size. Using this approach, we generated a set of 22 DSM with sizes ranging from 2.5 m2 to 4 m2 and an x-, y-resolution of 2 mm. The DSM were acquired over different roughness conditions representing ploughed, harrowed as well as crusted fields. For roughness characterization we calculated different roughness indices (RMS-height s, autocorrelation length l). In an extensive statistical investigation we show the behavior of the roughness indices for different acquisition sizes of the proposed method. Results indicate, compared to results from profiles generated out of the dataset, that using a three dimensional measuring device, the calculated roughness indices are more robust in their estimation. Results also indicate a strong directional dependency of the proposed roughness indices.